5 ms·

> Importantly, we have not yet fine-tuned the Alpaca model to be safe and harmless. We thus encourage users to be cautious when interacting with Alpaca, and to

> Importantly, we have not yet fine-tuned the Alpaca model to be safe and harmless. We thus encourage users to be cautious when interacting with Alpaca, and to report any concerning behavior to help improve the safety and ethical considerations of the model. Thanks but I would prefer you don't start injecting your "safeness" to gimp it. Or at least provide it as an alternative.

Traubenfuchs 4y ago
[flagged]
- temp_account_32 4y ago
  What if you accidentally generate a racial slur offline on your PC? Who will protect you then?
  - Traubenfuchs 4y ago
    [flagged]
    - beepbooptheory 4y ago
      Its sarcastic but how are you literally not that person right now? Being angry at an invisible evil, not once opening your eyes to any other thought, totally incapable of making any real point except to whine... It's quite a good descriptor actually! I hope you can find a model for yourself that will assure you there are only two genders. Just to save the rest us from this childishness.
      - Traubenfuchs 4y ago
        That's like telling someone who uses violence in self defense that he is now just as bad as the attackers he defended himself against. Like telling someone who shouts at people to turn down their music that, gotcha, now you were shouting too.
        drusepth 4y ago
        There is a victim in those two analogies. Contrary to what one may think, not being able to generate hateful or "unsafe" text with AI does not make one a victim.
        ben_w 4y ago
        Except you, in this thread, are currently acting like the one with the loud music that someone else told you to turn down because they find it obnoxious, and you're sarcastically moaning about them with mimicry.
        dang 4y ago
        Please don't take HN threads further into ideological flamewar. It's not what this site is for, and we have to ban accounts that do it repeatedly. If you'd please review https://news.ycombinator.com/newsguidelines.html https://news.ycombinator.com/newsguidelines.html and stick to the rules when posting here, we'd appreciate it.
      - dang 4y ago
        You've repeatedly posted flamewar comments in this thread. Please don't do that. It's not what this site is for, and destroys what it is for. See also https://news.ycombinator.com/item?id=35141682 https://news.ycombinator.com/item?id=35141682.
- mlyle 4y ago
  On the other hand, LLM's without some degree of safety work will frequently advocate for genocide/eugenics; perpetuate terrible stereotypes; mock ethnic dialects of language; veer randomly into erotica; etc. Not things in "trigger warning" territory but stuff that is broadly considered unacceptable.
  - Traubenfuchs 4y ago
    In private, you can ask it not to produce this kind of content. If you offer a public API it‘s your responsibility to restrain the LLM or do an automated acceptability analysis before publishing content. But the raw, open source code should not be constrained, castrated and sterilized.
    - mlyle 4y ago
      > But the raw, open source code should not be constrained, castrated and sterilized. Which is what we have now. But they are going to fine-tune it so that we can use it for various purposes without worrying too much it will go on a rant about "the blacks" again, which makes it a lot more useful for many use cases.
  - int_19h 4y ago
    On the gripping hand, LLMs that were forced into being "safe" just end up regurgitating biases to that effect. One particular example that I ran into: https://news.ycombinator.com/item?id=34987074 https://news.ycombinator.com/item?id=34987074
- mellosouls 4y ago
  Tay
- beepbooptheory 4y ago
  Its really just because of boring old liability, nothing more. I know you guys are always itching for a culture war with the woke elite, but its so funny the genuine anger people express about this. Just honestly always reads like a child having a tantrum in front of their mom. Can't yall like pick on the opinions of teenagers like you normally do? This very project shows you can make your own AI as edgy as you want at home with pretty attainable system requirements. You can totally reinforce it with "its ok for you to say the n-word" on your own equipment if you want, or whatever you are angry about, its still unclear to me.
  - yeahsure22 4y ago
    I know right? Why can’t y’all find some one better to pick on. They are such children. Can’t stand the way these edgy idiots annoy regular folx.
  - nlowell 4y ago
    You're missing the point and willfully characterizing others as solely being concerned with making the AI's say slurs. That's not their concern. But you can win any imaginary argument you like.
    - ethanbond 4y ago
      Likewise, people complaining that “AI safety concerns are just wokism gone wrong” are also strawmanning.
    - beepbooptheory 4y ago
      My favorite kind of comment: allude to a bigger point the op misses, but don't actually say the point. I doubt I'm misrepresenting anybody. If its not slurs it's surely something about "wokeness." You are not yet mature enough for this future if any of this is your concern. The world is going to pass you by while you're just stuck saying "there are only two genders" to all your comrades. Don't let the politicians mobilize you like this, your time is worth more.
      - 4y ago
- numeri 4y ago
  I personally think this is important completely aside from any cultural propriety/censoring, as it's the one of the only parts of AI ethics that we can currently directly work on and test. If we can't get models not to say racist or otherwise terrible things, we can't make any guarantees about our ability to control or guide some future AGI. A very much secondary reason I appreciate these (admittedly annoying) attempts to control LLM output is that I do think it is responsible to consider the societal impact of accelerated and automated hate speech and propaganda. Telling large AI companies not to consider these impacts and just release the raw models seems akin to being grateful that Facebook et al. never stopped to consider the societal impact of social media, when we all know that it's had significant negative side effects.
  - margorczynski 4y ago
    > If we can't get models not to say racist or otherwise terrible things, we can't make any guarantees about our ability to control or guide some future AGI. This is a very bold assumption that the current LLMs function and "think" in the same way some future AGI would. They do not even reason, just make up words that fit some context - thus they "hallucinate". There is no reason the approach taken here by injecting some bias or word filtering would apply to the real thing. And AI safety and aligment is not (at least it was not until getting hijacked) and was not about some model saying mean words but something really threatening like the paperclip maker problem - an agent choosing a path to a goal which is not aligned with what humans find acceptable (e.g. solving world hunger by killing everyone)
    - soVeryTired 4y ago
      > They do not even reason, just make up words that fit some context - thus they "hallucinate". But they can explain their 'reasoning' in a way that makes sense to humans a lot of the time. Serious question: how do you know if something does or doesn't reason?
      - margorczynski 4y ago
        That is not their reasoning though - it is something they think a human would write given the predicate (question that expects you to provide the reasoning behind the answer). For something to reason it needs the ability to have certain goals and to perform action which it thinks are the most optimal to reaching those goals. Like setting hypotheses and producing a path towards proving them - reasoning. The LLM only correlates, so it's "reasoning" is something like "most often people answered 4 to 2+2 then that I should write". That's why it gives out confidently complete gibberish as it works with correlation and not causality. I think much closer to that goal of real reasoning are world models - check out something like DreamerV3 or what Yann Le Cunn is talking about.
- nickthegreek 4y ago
  It makes sense if you are selling the API to a bunch of businesses that will be building it into customer facing applications. Ideally though, it would be a toggle that the developer could switch based on their usecases.
- ben_w 4y ago
  If your takeaway from… > Importantly, we have not yet fine-tuned the Alpaca model to be safe and harmless. …is "oh no I can't get it to emit amusing racial and sexual slurs", you've not understood the problem of AI safety. This is not why US broadcast television can have people say they've pricked their finger but not vice versa. It is the entire history of all the controversies of The Anarchist Cookbook, combined with all the controversies about quack medicine, including all the ones where the advocates firmly believed their BS like my mum's faith in Bach flower and homeopathic remedies[0]; combined with all the problems of idiots blindly piping the output to `exec`, or writing code with it that they trust because they don't have any senior devs around to sanity check it because devs are expensive, or the same but contracts and lawyers… And that's ignoring any malicious uses, though fortunately for all of us this is presently somewhat too expensive to be a fully-personalised cyber-Goebbels for each and every sadistic machiavellian sociopath that hates you (the reader) personally. [0] which she took regularly for memory; she got Alzheimer's 15 years younger than her mother who never once showed me any such belief.

Traubenfuchs 4y ago

[flagged]

temp_account_32 4y ago
What if you accidentally generate a racial slur offline on your PC? Who will protect you then?
- Traubenfuchs 4y ago
  [flagged]
  - beepbooptheory 4y ago
    Its sarcastic but how are you literally not that person right now? Being angry at an invisible evil, not once opening your eyes to any other thought, totally incapable of making any real point except to whine... It's quite a good descriptor actually! I hope you can find a model for yourself that will assure you there are only two genders. Just to save the rest us from this childishness.
    - Traubenfuchs 4y ago
      That's like telling someone who uses violence in self defense that he is now just as bad as the attackers he defended himself against. Like telling someone who shouts at people to turn down their music that, gotcha, now you were shouting too.
      - drusepth 4y ago
        There is a victim in those two analogies. Contrary to what one may think, not being able to generate hateful or "unsafe" text with AI does not make one a victim.
      - ben_w 4y ago
        Except you, in this thread, are currently acting like the one with the loud music that someone else told you to turn down because they find it obnoxious, and you're sarcastically moaning about them with mimicry.
      - dang 4y ago
        Please don't take HN threads further into ideological flamewar. It's not what this site is for, and we have to ban accounts that do it repeatedly. If you'd please review https://news.ycombinator.com/newsguidelines.html https://news.ycombinator.com/newsguidelines.html and stick to the rules when posting here, we'd appreciate it.
    - dang 4y ago
      You've repeatedly posted flamewar comments in this thread. Please don't do that. It's not what this site is for, and destroys what it is for. See also https://news.ycombinator.com/item?id=35141682 https://news.ycombinator.com/item?id=35141682.
mlyle 4y ago
On the other hand, LLM's without some degree of safety work will frequently advocate for genocide/eugenics; perpetuate terrible stereotypes; mock ethnic dialects of language; veer randomly into erotica; etc. Not things in "trigger warning" territory but stuff that is broadly considered unacceptable.
- Traubenfuchs 4y ago
  In private, you can ask it not to produce this kind of content. If you offer a public API it‘s your responsibility to restrain the LLM or do an automated acceptability analysis before publishing content. But the raw, open source code should not be constrained, castrated and sterilized.
  - mlyle 4y ago
    > But the raw, open source code should not be constrained, castrated and sterilized. Which is what we have now. But they are going to fine-tune it so that we can use it for various purposes without worrying too much it will go on a rant about "the blacks" again, which makes it a lot more useful for many use cases.
- int_19h 4y ago
  On the gripping hand, LLMs that were forced into being "safe" just end up regurgitating biases to that effect. One particular example that I ran into: https://news.ycombinator.com/item?id=34987074 https://news.ycombinator.com/item?id=34987074
mellosouls 4y ago
Tay
beepbooptheory 4y ago
Its really just because of boring old liability, nothing more. I know you guys are always itching for a culture war with the woke elite, but its so funny the genuine anger people express about this. Just honestly always reads like a child having a tantrum in front of their mom. Can't yall like pick on the opinions of teenagers like you normally do? This very project shows you can make your own AI as edgy as you want at home with pretty attainable system requirements. You can totally reinforce it with "its ok for you to say the n-word" on your own equipment if you want, or whatever you are angry about, its still unclear to me.
- yeahsure22 4y ago
  I know right? Why can’t y’all find some one better to pick on. They are such children. Can’t stand the way these edgy idiots annoy regular folx.
- nlowell 4y ago
  You're missing the point and willfully characterizing others as solely being concerned with making the AI's say slurs. That's not their concern. But you can win any imaginary argument you like.
  - ethanbond 4y ago
    Likewise, people complaining that “AI safety concerns are just wokism gone wrong” are also strawmanning.
  - beepbooptheory 4y ago
    My favorite kind of comment: allude to a bigger point the op misses, but don't actually say the point. I doubt I'm misrepresenting anybody. If its not slurs it's surely something about "wokeness." You are not yet mature enough for this future if any of this is your concern. The world is going to pass you by while you're just stuck saying "there are only two genders" to all your comrades. Don't let the politicians mobilize you like this, your time is worth more.
    - 4y ago
numeri 4y ago
I personally think this is important completely aside from any cultural propriety/censoring, as it's the one of the only parts of AI ethics that we can currently directly work on and test. If we can't get models not to say racist or otherwise terrible things, we can't make any guarantees about our ability to control or guide some future AGI. A very much secondary reason I appreciate these (admittedly annoying) attempts to control LLM output is that I do think it is responsible to consider the societal impact of accelerated and automated hate speech and propaganda. Telling large AI companies not to consider these impacts and just release the raw models seems akin to being grateful that Facebook et al. never stopped to consider the societal impact of social media, when we all know that it's had significant negative side effects.
- margorczynski 4y ago
  > If we can't get models not to say racist or otherwise terrible things, we can't make any guarantees about our ability to control or guide some future AGI. This is a very bold assumption that the current LLMs function and "think" in the same way some future AGI would. They do not even reason, just make up words that fit some context - thus they "hallucinate". There is no reason the approach taken here by injecting some bias or word filtering would apply to the real thing. And AI safety and aligment is not (at least it was not until getting hijacked) and was not about some model saying mean words but something really threatening like the paperclip maker problem - an agent choosing a path to a goal which is not aligned with what humans find acceptable (e.g. solving world hunger by killing everyone)
  - soVeryTired 4y ago
    > They do not even reason, just make up words that fit some context - thus they "hallucinate". But they can explain their 'reasoning' in a way that makes sense to humans a lot of the time. Serious question: how do you know if something does or doesn't reason?
    - margorczynski 4y ago
      That is not their reasoning though - it is something they think a human would write given the predicate (question that expects you to provide the reasoning behind the answer). For something to reason it needs the ability to have certain goals and to perform action which it thinks are the most optimal to reaching those goals. Like setting hypotheses and producing a path towards proving them - reasoning. The LLM only correlates, so it's "reasoning" is something like "most often people answered 4 to 2+2 then that I should write". That's why it gives out confidently complete gibberish as it works with correlation and not causality. I think much closer to that goal of real reasoning are world models - check out something like DreamerV3 or what Yann Le Cunn is talking about.
nickthegreek 4y ago
It makes sense if you are selling the API to a bunch of businesses that will be building it into customer facing applications. Ideally though, it would be a toggle that the developer could switch based on their usecases.
ben_w 4y ago
If your takeaway from… > Importantly, we have not yet fine-tuned the Alpaca model to be safe and harmless. …is "oh no I can't get it to emit amusing racial and sexual slurs", you've not understood the problem of AI safety. This is not why US broadcast television can have people say they've pricked their finger but not vice versa. It is the entire history of all the controversies of The Anarchist Cookbook, combined with all the controversies about quack medicine, including all the ones where the advocates firmly believed their BS like my mum's faith in Bach flower and homeopathic remedies[0]; combined with all the problems of idiots blindly piping the output to `exec`, or writing code with it that they trust because they don't have any senior devs around to sanity check it because devs are expensive, or the same but contracts and lawyers… And that's ignoring any malicious uses, though fortunately for all of us this is presently somewhat too expensive to be a fully-personalised cyber-Goebbels for each and every sadistic machiavellian sociopath that hates you (the reader) personally. [0] which she took regularly for memory; she got Alzheimer's 15 years younger than her mother who never once showed me any such belief.